Monday, January 17, 2011

How do you get at the question "What is a language?"

Other than a dialect with an army and a navy, I mean.

"What is English?"

Well, um, it's a language, a system with its own rules of morphology, phonology, orthography, syntax, etc., which can all be described by very specific references to...well, to itself, I suppose. That is interesting. One begins to understand the distinction between the "traditional" study of language-as-system and the "less traditional" study of language-in-use. Where does the system come from? Who makes it? While we all seem to have the innate capacity to understand and do language, we don't really map out the system ourselves. Nobody has the blueprints (despite what L'Academie might wish), yet we are constantly building. Bakhtin tells us that no one short of God has made an original utterance (and that I assume he means metaphorically).

The thing is, then, that language is made and re-made and grows and changes constantly. Maybe the changes take a while -- half a century to get the current connotations of, say, gay in English or tongzhi in Mandarin --but language is what it is because we keep on making it what it is. (Why? I have no idea what causes us to keep changing it. Maybe it has something to do with boredom.)

What I'm getting at is, I used to think that the most important thing for getting at a definition of a language -- or specifically, a variety of English -- was a hardcore analysis of the stuff of language itself -- linguistic or rhetorical or discourse analysis. What I think I think now (not a typo), though, is that those kinds of analysis are very good for some things, but they're not necessarily ideal for answering the question "what is this language?" (Well, there's also corpus analysis, which I used to think was overly sciencey and academic, but which I am more and more interested in since it is based on real samples of language-in-use.)

People -- everyday schlubs like you and me -- are the ones who make language, and the ones who decide what it is. So one of the best ways to figure out what a language is -- and this is going to sound so simple and naive as to be almost absurd -- is to ask people what it is.

Is this English?
Is this good writing?
Is this standard English?
Is this a well-written sentence?
Is this grammar OK?
What does this mean? Would you write it that way if it was your essay?
Why not?


Xiaoye said...

Two questions - 1. In what ways what you achieve through corpus analysis different from what you achieve by viewing language as a system? The same question for using your "simple" sampling approach. 2. How to answer the question, "What is language," when taking a language-as-practice perspective?

Joel said...

Sorry I haven't replied -- these are hard questions!

I don't know a lot about corpus linguistics but it seems like using a corpus as a way to look at "actual usage" (however small the sample) is satisfying because you don't have to make as many anecdotal inferences or assumptions. But the results of any study using a corpus will be very limited, I assume, since there could always be other samples of language that could differ from your original corpus.

The second question is much harder, and that is something I am wrestling with when it comes to theory. Saraceni suggests that talking about varieties in the traditional sense (varieties as differing systems?) isn't even necessary, and that functions and attitudes/ideologies of English in different contexts are more important now. I'm not sure I agree, but it leaves me wondering what to do with actual samples of any variety of English...

The more I think about this the more I get confused...